It is becoming increasingly popular to record video during law enforcement activities, such as by using wearable cameras, in-car dash cameras, and/or the like. Even though large amounts of video may be collected using these techniques, the videos are often subject to disclosure under the Freedom of Information Act (FOIA) or similar laws or policies. To comply with such laws while maintaining the privacy of depicted individuals, sensitive information is commonly redacted from the videos before disclosure.
Public agencies are becoming inundated with FOIA requests once video capture is rolled out. Using existing tools, it often takes several hours to redact a few minutes of video, as video data must be reviewed and redacted for minors, undercover officers, innocent bystanders, and other sensitive information. Furthermore, audio data also must be redacted for names, addresses, and any other sensitive information. It has become such a huge burden that some public agencies have publicly stated that they are considering ending video capture projects as they cannot support the requests. What is needed are methods and systems that help reduce the amount of time needed to redact videos in response to public records requests.